Screening and molecular characterization of cholesterol-lowering lactic acid bacteria isolated from African oil palm wine (Elaeis guineensis) and corn beer.

Background: High serum cholesterol is a risk factor for cardiovascular disease (CVD), leading to many fatalities. Probiotic supplementation improves health benets by lowering serum cholesterol. African oil palm (Elaeis guineensis) wine and corn beer have been described as potential sources of probiotic bacteria, but their cholesterol lowering activity has not been extensively investigated. Therefore in this study, we isolated and characterized lactic acid bacteria with probiotic and cholesterol lowering properties from these sources. Methods: A total of fty lactic acid bacteria (LAB) were isolated from palm wine and corn beer using de Man Rogosa and Sharpe (MRS) agar by pour plate method. Catalase negative and Gram positive colonies were considered as presumptive LAB. For probiotic properties and selection, acid and bile salt tolerance were evaluated based on viable colony count on MRS agar and cholesterol assimilation from culture media was performed by spectrophotometry. The data analysis was performed by One Way Analysis of Variance and p value< 0.05 was considered signicant. The isolates with excellent cholesterol assimilation properties were selected and identied with species of morphology and biochemical analysis using API 50 CHL kit and genotypic identication by sequencing of 16 S rRNA gene. Results: Fifty isolates had morphological features similar to LAB. They were catalase negative and Gram positive colonies. Among the fty isolates, ve were considered as acid and bile tolerant (viable count exceeded 6logCFU/ml), with two isolates having excellent cholesterol assimilation property (>80%). The result obtained using API 50 CHL kit did not agree with the genotypic identication which was used to conrm the identity of the isolates as most closely related to Lactobacillus brevis ATCC (Accession No.NR116238) and Enterococcus faeccium NBRC (Accession No. NR113904). Conclusions: The two strains isolated from traditionally fermented drinks in Buea can be considered as potential probiotics. They can therefore be used to formulate food supplement needed to lower cholesterol in hypercholesterolemia patients.


Background
Cholesterol is a naturally occurring substance in the human body that plays vital roles in cell wall biosynthesis and hormone production. The liver and the intestinal mucosa are capable of synthesizing about 80% of de novo (endogenous) cholesterol needed in the body [1], requiring a small amount (20%) of dietary (exogenous) cholesterol. However, when dietary cholesterol becomes higher, it can lead to a condition known as hypercholesterolemia (elevated blood cholesterol). This causes the accumulation of cholesterol deposits in blood vessels, leading to blockage or narrowing of arteries which supply blood to vital organs like the heart and brain. Hypercholesterolemia is the major risk factor for cardiovascular diseases (CVD) [2]. It has been predicted to remain the single leading cause of death for the next ten years, killing approximately 23.6 million people [3]. More than 80% of CDV deaths occur in low and middle-income countries, affecting men and women at the same rates [4]. A recent study carried out in Cameroon showed that deaths related to CDV (in-hospital case-fatality) were 15.8% [5].
The use of low-fat diet so far is the most effective means of lowering elevated serum cholesterol [6]. This remedy however has limitations due to the unavailability of such diets and compliance of the consumer [7]. Probiotic bacteria appear as the most promising alternative. They are living organisms that confer a health bene t to the host when administered in adequate amount [8]. Lactic acid bacteria (LAB) remain the most common types of microbes used as probiotics, although certain yeasts are also used [9][10].
Fermented drinks like palm wine and corn beer have been noted as potential niches that harbor a rich consortium of microorganisms due to the presence of simple sugars used as a substrate for growth. A study carried out by Chandrasekhar et al. [11] and Parveens and Ha z, [12] showed that LAB are among the most predominant microorganisms in locally fermented drinks. Palm wine and corn beer are widely consumed beverages in Africa and Cameroon in particular due to their low cost and ready availability. In addition to this extensive consumption, there is very limited cognizance of the inherent health bene ts of the microorganisms that are predominant in these beverages [13]. The cholesterol assimilation property among others stands out as one of the most signi cant functional properties of LAB which can be used to lower cholesterol levels in hypercholesterolemia patients.
However, little is known about the cholesterol lowering e cacy of probiotic LAB from locally harvested palm wine and traditionally processed corn beer in Cameroon. In this study, we isolated and characterized LAB with probiotic properties and their ability to lower cholesterol in vitro.

Sample collection
Sap from African oil palm (Elaeis guineensis) was procured from palm wine tappers. Sap collection was done following the destructive method described by Onuche et al. [14]. Brie y, the tree was cut down and a cavity created by digging into the soft meristem of the tree trunk. A tube was inserted to make way for sap collection in a sterile plastic bottle. Corn beer was collected with the use of a sterile plastic bottle from corn beer vendors. African oil palm and maize plant were identi ed at the National Herbarium in Media preparation was performed following the manufacturer's instructions. Tenfold serial dilution was made by transferring 1 ml of each sample into 9 ml of peptone water. The pour plate method was used to enumerate bacteria cells (CFU/ml). Then, 0.1 ml from each dilution was transferred into sterile Petri dishes and covered with molten agar. The plates were incubated at 37 o C for 24 h. Repeated sub-culture by streaking on MRS agar was carried out to obtain pure colonies. Pure colonies were labeled with codes, with Pw for palm wine LAB, and Cb for corn beer LAB and Arabic numbers attributed starting from 1.
Preliminary identi cation involving colony morphology, Gram staining, and catalase tests were performed.

Probiotic properties of LAB
The major selection criteria used to determine the probiotic properties of LAB isolates were tolerance to low pH, bile salt tolerance, and in vitro cholesterol assimilation activity. All tests were performed in triplicates and the number of viable colonies in MRS agar plates was counted.
Tolerance to acid pH values Acid tolerance was evaluated following the method described by Guan at al. [15] with slight modi cations. Overnight cultures of bacteria cells were washed three times with PBS (pH 7.0) to remove impurities and centrifuged (Eppendorf centrifuge 5810 R, New York, USA) for 10 min at 4 o C at 5,000 rpm. The cell pellets were re-suspended in MRS broth adjusted to pH 2.0 and pH 3.0 (HI991001, Woonsocket, USA) using 3N HCl or NaOH. The cultures were then incubated at 37 o C for 24 h. Aliquots were taken after 0 h and 3 h, serially diluted. Samples taken at 0 h were used as the control. Isolates that exhibited nal counts ≥ 10 3 CFU/ml or ≥ 10 6 CFU/ml at low pH for 3 h, were considered to have moderate or good resistance, respectively. To perform enumeration, 1 ml of each of the suspensions was serially diluted up to the ten logarithmic fold and the viable microorganisms were counted in triplicates on MRS agar.

Bile tolerance
The bile salt resistance of selected isolates was determined by the method described by Argyri et al. [16] with minor modi cations. Overnight cultures of bacteria cells were washed three times with PBS (pH 7.0) to remove impurities and centrifuged (5,000 rpm for 10 min at 4 o C). The cell pellets were re-suspended in MRS broth containing 0.2 and 0.4% oxgall bile salts (sigma Aldrich, Germany). The cultures were then incubated at 37 o C for 24 h. Aliquots were taken after 0 h and 3 h, serially diluted, and plated on MRA agar. Samples taken at 0 h were used as the control. Resistance to bile salt was evaluated based on viable colony counts on MRS agar in triplicates after incubation at 37 o C for 0 and 3 h, re ecting the average time spent by food in the small intestine.
Cholesterol assimilation from culture media Based on the acid and bile tolerance of the selected strains, the ability of each strain to assimilate cholesterol in vitro was determined by a modi ed method described by Pereira and Gibson [17]. Bacteria strains were inoculated into tubes, each containing 10 ml of MRS broth, 0.4% bile salts, and 1% acid solution of cholesterol (Sigma-Aldrich, cat # C3045-5G, Germany). The cultures were incubated at 37°C for 24 h. After incubation, the cultures were centrifuged (5,000 rpm for 10 min at 4 o C) and the unutilized cholesterol estimated in the supernatant. This was carried out by spectrophotometry (Pharmacia LKB, England) at 540 nm and compared to the control as described by Ngongang et al. [18]. The percentage of cholesterol assimilation was determined by the equation established by Al-Sahel et al. [19] Where A is the percentage of cholesterol that remained with the pellet, B is the absorbance of the sample containing the cells, and C is the absorbance of the sample without cells.
Isolates having in vitro cholesterol assimilation properties were selected for biochemical identi cation using API 50 CHL assay.

Identi cation of LAB isolates
Phenotypic identi cation of LAB isolates using API 50 CHL kit Phenotypic identi cation of LAB isolates was performed by API 50 CHL (API kit, bioMérieux, France) assay. Puri ed LAB cultures were cultivated in 20 ml MRS broth incubated at 37°C overnight, after which they were washed and re-suspended in API ® 50 CHL medium (bioMerieux ® SA 69280, France). The turbidity of the suspensions was determined by the McFarland method according to the instructions provided by the manufacturer. Cell suspensions were transferred into API 50 CHL strip wells and overlaid with para n oil to create an anaerobic condition. The strips were incubated at 37°C. The results were read after 24 h and con rmed after 48 h. Fermentation of carbohydrates was indicated by a yellow color except for the esculine test (black). Color reactions were scored against a chart provided by the manufacture [20]. The results were analyzed with API WEB (bioMerieux) database version 5.0.
Genotypic identi cation of LAB isolates using 16 S rRNA gene sequencing Genomic extraction of the two strains of LAB was determined following the method described by Mulaw et al. [21] with some slight modi cations.

DNA extraction of LAB isolates
The genomic DNA was extracted from pure cultures of isolate Pw4 and Cb5. One ml of each pure liquid culture was centrifuged at 11,500 rpm for 10 min at 25 0 C (Eppendorf centrifuge 5810 R, New York, USA).
The supernatant was decanted and the cell pellets re-suspended into a tube containing 300 μl buffer (10mM Tris-HCl, pH 8.0; 50mM glucose, and 10mM EDTA) and 3 µl lysozyme (10 mg/ml). The pellets were lysed at 37 0 C for 60 min and vortexed every 5 min, followed by placing in ice every 5 min. Threefold (300) µl of lysis buffer and 3 µl RNAse were added to the mixture and incubated for 30 min and cooled on ice for 1 min. Then, 100 µl of 7.5 M solution of sodium acetate was added and vortexed for 25 seconds and centrifuged (Eppendorf centrifuge 5810 R, New York, USA) at 13,000 rpm for 10 min at 4 o C. The supernatant was transferred into a sterile tube, and 300 μl isopropanol was added and mixed gently.
The resulting mixture was centrifuged at 13,000 rpm for 10 min at 4 o C (Eppendorf centrifuge 5810 R, New York, USA). Isopropanol was carefully removed by the use of a sterile Eppendorf pipette without dislocating the DNA pellets. The tubes were air-dried by inverting them on sterile lter paper. The DNA was washed by adding 400 µl of 70% ethanol and centrifuged at 5,000 rpm for 2 min at ambient temperature.
The sediments were dried at 37 o C for 10 min and nally dissolved in 30 µl TE buffer and stored at -20 o C for further study.

Gel electrophoresis
Two µl of each ampli cation mixture was subjected to electrophoresis in 1.5% (w/v) agarose gels in 0.5 x TAE buffer for 1 h at100 V. The DNA molecular mass marker (250 to 10000 bp) from inqaba biotech, South Africa was used as the standard. After electrophoresis, the gels were stained with ethidium bromide, washed, and photographed with UV transilluminator (Bio-Rad, Hercules, CA, USA). The partial 16S rRNA sequence analysis of the PCR products was sequenced by inqaba biotech, South Africa. The sequences obtained were compared using BLAST (basic local alignment search tool) and submitted to the GenBank sequence database for accession numbers [22]. A phylogenetic tree was constructed using MEGA 10 software to reduce all positions containing gaps and missing data in the trail sequence in order to evaluate the evolutionary relationship of Pw4 and Cb5 and their close relations.

Statistical analysis
All the tests were performed in triplicate, and the results were expressed as mean ± standard deviation.

Results
Isolation of lactic acid bacteria wine and 18 from corn beer all sourced from Buea, South West Region of Cameroon. Preliminary identi cation of colonies carried out on the basis of cell morphology, microscopic examination and biochemical tests revealed smooth, oval, and cream white colonies on MRS agar plate. However, only ten colonies were catalase negative and Gram positive under microscopic examination and were considered as presumptive LAB. Table 1 shows the preliminary identi cation of the isolates.

Resistance to acid
The viable counts of all isolates ranged from 3.9-7 log CFU/ml after 3 h of exposure at pH 2.0. This ability of the strains to survive the acidic pH value after 3 h of incubation at 37°C is presented in gures 1a and 1b. Isolates Pw1, Pw4, Cb1, Cb3, Cb4, Cb5 and Cb6 had viable counts greater than 6 log CFU/ml, which was signi cantly different (p<0.05) from the control (pH2.0 at 3 h). On the contrary, isolates Pw2, Pw3 and Cb2 had viability below 5 log CFU/ml which was not signi cantly different (p>0.05) when compared to the control. However, when the pH value was increased to 3.0 for 3 h, the viability of all isolates increased in the range of 4.3-7 log CFU/ml. The viable counts of isolates Pw1, Pw3, Pw4, Cb1, Cb3, Cb4, Cb5, Cb6 exceeded 6 log CFU/ml (p<0.05 at pH3.0 for 3 h). Contrariwise, the viability of isolates Pw2 and Cb2 fell below 5 log CFU/ml (p>0.05). In total, isolate Pw1, Pw4, Cb1, Cb3, Cb4, Cb5 and Cb6 maintained a high viable count despite the change in pH values.

Resistance to bile salt
The viability of all the isolates ranged from 3.9-7.5 log CFU/ml after 3 h of exposure to 0.2 % bile salt concentration. The viable count of isolate Pw1, Pw4, Cb1, Cb2, Cb4, Cb5 and Cb6 was higher than 6 log CFU/ml which was signi cantly different from the control (p<0.05). On the other hand, isolates Pw2, Pw3 and Cb3 had viability below 4.5 log CFU/ml. When the bile salt concentration was increased to 0.4 % after 3 h, the viability of all the isolates fell in the range of 3.1-7 log CFU/ml. The viability of Pw2, Pw4, Cb1, Cb4, Cb5 and Cb6 were greater than 6 log CFU/ml which were not signi cantly greater than the control (p>0.5). On the contrary, isolates Pw1, Pw3, Cb2 and Cb3 had viable counts below 5 log CFU/ml which were signi cantly different from the control (p<0.5).
In total, isolates Pw4, Cb1, Cb4, Cb5 and Cb6 maintained good viability irrespective of the change in bile salt concentration. This ability of the strains to survive the bile salt concentrations after 3 h of incubation at 37°C is presented in gures 2a and 2b.
Cholesterol assimilation from culture media The amount of cholesterol assimilated in-vitro ranged from 28 -89 % after 24 h of incubation in the presence of bile salt ( gure 3). Seven strains assimilated cholesterol at a variable extent with Pw1, Pw4, Cb4 and Cb5 which displayed excellent assimilation property (>75%). Contrariwise, Pw3 and Cb3 had poor cholesterol (<45%) uptake and did not grow well in the medium. Strains Pw4 and Cb5 were selected for further studies.

Identi cation of LAB isolates
Phenotypic identi cation of LAB isolates using API 50 CHL kit assay Results from API 50 CH test kits and API web identi ed the two LAB isolates (Pw4) from palm wine Lactobacillus plantarum with similarity 99.6 % and (Cb5) from corn beer Lactobacillus. pentosus 99.5% (Table 2). There was an insigni cant variation in the utilization of carbohydrates sources of the API CHL Page 9/17

Genotypic identi cation of LAB isolates
The genotype identi cation of DNA using universal primer showed clear bands of isolates on agarose gel with approximate molecular weight 15000 bp. Results obtained using BLAST identi cation and MEGA 10 software revealed close similarity of 97.4% of Pw4 to Lactobacillus brevis strain ATC (Accession no. NR116238) and 99.54% of Cb5 to Enteroccocus faecium strain NBRC (Accession no.NR113904).
Phylogenetic trees of these selected isolates based on 16S RNA gene sequences are presented in Figures  4a and 4b.

Discussion
Locally harvested palm wine and traditionally processed corn beer have been exploited as suitable sources for screening various LAB [23]. In Cameroon, palm wine and corn beer are widely consumed with little or no knowledge of the presence of a predominant LAB population [13]. Out of the fty samples collected and grown on MRS agar plates, only ten isolates had morphological (microscopic and macroscopic), biochemical features similar to LAB (Table 1) as reported by Bennani et al. [24]. Four of these isolates were sampled from palm wine coded as Pw1, Pw2, Pw3, and Pw4 and six from corn beer coded Cb1, Cb2, Cb3, Cb4, Cb5 and Cb6. This implies that LAB cultures were predominant in corn beer when compared to palm wine. Fresh palm wine has been reported to have a rich population of microbial load than corn beer [25]. This causes rapid bioconversion of sugar present in palm wine to alcohol during fermentation. However, in this study, LAB cultures were isolated more (60%) from corn beer than from palm wine (40%) (Table1). This indicated that a high microbial population is not synonymous to a high LAB presence. The nding in this study concurs with that of Nwachukwu et al. [26] who successfully isolated LAB from fermented samples.
Probiotic LAB have been shown to have inherent health bene ts and approved for human consumption primarily because they have acquired a status of GRAS (Generally Regarded as Safe) and QPS (Quali ed Presumption of Safety) by the American Food and Drug Agency (FDA) and the European Food Safety Authority (EFSA), respectively [27]. One of such functional property proven to be more effective in hypercholesterolemia patients involves cholesterol assimilation in the liver and ileum. Recent studies have proven that probiotic LAB are able to assimilate cholesterol to a signi cant margin [28][29]. Gilliland et al. [30] were the rst to show that in vitro e ciency of lactobacilli could be directly associated with their ability to assimilate cholesterol. As a result, in-vitro removal of cholesterol by lactobacilli has been consistently used as a screening tool for the selection of probiotic strain. In this study, in vitro cholesterol assimilation assay was performed to assess the cholesterol assimilation property of ten LAB. For e cient cholesterol assimilation, 0.4 % bile salt was added to the medium and grown under aerobiosis as described by Anandharaj et al. [31]. The results obtained ( gure 3) showed two isolates (Pw4 and Cb5) with the best cholesterol assimilation ability (>80%). This implies that isolate Pw4 and Cb5 can e ciently be used to assimilate cholesterol in vitro. A similar study by Anila et al. [32] and Malakar et al. [33] revealed that L. brevis PLA 7, L. brevis PLA 14 and Enteroccocus faecium BASTUS 4 and 5 respectively isolated from fermented beverages were able to assimilate cholesterol signi cantly in vitro. A possible mechanism of cholesterol assimilation of Pw4 and Cb5 was the cell wall binding and incorporation of cholesterol within their phospholipid layer as described by Ooi and Lion [34]. However, Ramasamy, [35], highlighted that bile salt hydrolytic activity (BSH) of probiotics stands as one of the most signi cant mechanisms for cholesterol removal and has been proposed by many researchers as a prerequisite for probiotic selection. Evaluating the in vitro cholesterol assimilation was signi cant to extrapolate a similar trend during in vivo studies. A study carried out by Ngongang et al. [18] revealed alike trend during cholesterol assimilation in MRS culture medium as well as in albino rats.
In spite of the high cholesterol assimilation ability of LAB, they must overcome the stressful conditions of acid (stomach pH) and bile (duodenum) before initiating their health bene t (cholesterol lowering effect) in the ileum. Acid and bile tolerance is therefore regarded as a prerequisite for any LAB to be considered as a probiotic [36]. The survival of LAB in the stomach is important because the pH of the stomach (HCl acid) can be as low as 1.5 (without food), inhibiting metabolic activities thereby retarding the viability of Lactobacillus sp. The transition of this acidic medium (3 h) is vital for probiotics to initiate their bene cial effects on the distal part of the gut [37]. In this study, ten presumptive LAB were exposed to pH 2 and 3 for 3 h to assess their ability to withstand stressful conditions. Among the ten isolates, seven had viable counts reaching 6log CFU/ml irrespective of the pH change. This implied that these isolates were able to tolerate acidic conditions at pH 2.0 and pH 3.0 and were therefore considered acid tolerant. Similar ndings have been reported by Somashekaraiah et al. [38], who reported excellent acid tolerant properties of Enterococcus faecium (MH748610-MYSN ) and Lactobacillus brevis (MH748630-MYSN) isolated from neera ( a naturally fermenting coconut palm nectar). The increase in viability in pH 3.0 compared to pH 2.0 was similar to the results obtained by Vanniyasingam et al. [39].
Evaluating tolerance to bile is important because probiotic bacteria have to transit the duodenum (where bile acid is present) before initiating cholesterol assimilation in the ilium [40]. Bile salt is synthesized in the liver from cholesterol and stored in the gall bladder from where it is secreted into the duodenum in a conjugated form during fat metabolism. The physiological concentrations of human bile salts range from 0.3 to 0.5% [41]. In this study, the viable count of the isolates was determined after exposure to 0.2 and 0.4 % bile salt after 3 h of incubation. Seven strains were found to tolerate 0.2 % bile salt, while six strains were able to tolerate 0.4 % bile salt. This implies that high bile concentration affects the viability of isolates. This is because disruption of cellular homeostasis causes the dissociation of the lipid bilayer and integral protein of their cell membranes, resulting in bacterial content leakage and cell death [42]. A similar result showed that Enterococcus faecium (MH748610- MYSN 18) and Lactobacillus brevis (MH748630-MYSN) isolated from some fermenting coconut palm nectar had bile salt tolerance of more than 80% [38]. Bile salt hydrolytic activity which is associated with the capacity to eliminate cholesterol from the intestinal environment has been shown to be an important mechanism for bile tolerance [43].
Conventional phenotypic methods alone are insu cient for the identi cation of LAB. Reliability can be achieved by the use of molecular techniques. The 16S rRNA gene is found to be a powerful tool for appreciating genetic variability among different species [44], which allows for the design and use of universal primers to discover and classify organisms into a wide range of taxa. Mulaw et al. [45] and Dowarah et al. [46] have revealed the strain level identi cation of diverse LAB with potent probiotic properties isolated from traditionally fermented foods and other substrates using phylogenetic estimation of 16S rDNA genes. In this study, the results obtained from the two identi cation methods did not agree (Tables 2 and 4). However, discrepancies between sequencing and phenotypic (API 50CH kit) methods have been reported for LAB and other bacteria [47]. The API 50 CHLassay is intended uniquely for the identi cation of those species included in the database. As a result, the strains not found in the identi cation table provided by bioMérieux SA, were successfully identi ed using the 16S rRNA gene sequencing. Moreover, a study by Bağder et al. [48] compared the results of the API 50 CHL test with 16 S rRNA results and found that the API test did not give reliable identi cation results, with only 71 out of 152 tested isolates in agreement.
In this study, a phylogenetic tree for each gene was constructed by the Maximum Likelihood method and minimum evolution method to access the evolutionary relationship between local samples (Pw4 and Cb5) and the highest query cover above 90% of samples [49]. The sequences of Pw4 and Cb5 showed similarity of 97.4% at 97% query coverage to Lactobacillus brevis strain ATCC 14869 (Accession no. NR116238) and 99.54% at 97% query coverage to Enteroccocus faecium strain NBRC 100486 (Accession no.NR113904) respectively. Taken into consideration these two identi cation methods in this study, the results obtained a rms the need of using molecular methods for typing newly isolated microorganisms.

Conclusion
Lactobacillus brevis strain ATCC 14869 (Accession no. NR116238) and Enteroccocus faecium strain NBRC 100486 (Accession no.NR113904) isolated from palm wine and corn beer in Buea, have good acid and bile tolerance and can be used in cholesterol assimilation especially for patients displaying symptoms of high serum cholesterol. The properties of these LABs identi ed, can permit them overcome the stressful acid and bile salt conditions in the stomach and duodenum respectively, to provide health bene ts in the gut.